A.I. – One of the most misused terms in our industry
人工智能 – 我们行业中最被误用的术语之一

So by now if you have read any of my previous articles you probably know my feelings on companies who use the term A.I. by now. I thought I would take some time from my hospital bed (I am currently recovering from an operation) to clarify what exactly I mean and why.
所以，如果你读过我以前的任何文章，你可能现在就知道我对使用人工智能这个词的公司的看法了。我想我会从医院的病床上抽出一些时间（我目前正在从手术中恢复）来澄清我到底是什么意思以及为什么。

Lets start by giving it its full name- Artificial Intelligence.
让我们先给它起个全名——人工智能。

From the dictionary- 从字典中-

Artificial 假

Made or produced by human beings rather than occurring naturally, especially as a copy of something natural.
由人类制造或生产，而不是自然发生，尤其是作为自然事物的复制品。

Intelligence 情报

The ability to acquire and apply knowledge and skills.
获取和应用知识和技能的能力。

And both together 两者结合在一起

The theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.
计算机系统的理论和发展能够执行通常需要人类智能的任务，例如视觉感知、语音识别、决策和语言之间的翻译。

The person that coined this term was of course, the great Alan Turing. He wrote a paper called Computing Machinery and Intelligence in 1950. I know youngsters around here probably have never heard of this – but I remember reading it in university (although is was not on the curriculum) and staying up all night very excited about it. Why? Because its still relevant today. Regardless of how much computing has come on since 1950 we are still essentially in exactly the same place. You see, processors are the problem. Yes- you heard me. A processor can only do one thing at a time. We don’t notice that during our daily tasks, as it does that one thing very fast indeed these days. But never the less, we still can only do one thing at a time in computing. Now before we get mad, I am talking about kit available to me and other normal people, and I don't have a quantum computer to hand.
创造这个词的人当然是伟大的艾伦·图灵。他在 1950 年写了一篇名为《计算机与智能》的论文。我知道这里的年轻人可能从未听说过这个——但我记得在大学里读过它（尽管它不在课程中），并且整晚都对它感到非常兴奋。为什么？因为它在今天仍然适用。无论自 1950 年以来出现了多少计算，我们基本上仍然处于完全相同的位置。你看，处理器是问题所在。是的——你听到了。处理器一次只能执行一项操作。在我们的日常任务中，我们没有注意到这一点，因为现在它确实非常快地完成一件事。但无论如何，在计算领域，我们仍然只能一次做一件事。现在，在我们生气之前，我说的是我和其他正常人可用的工具包，而我手头没有量子计算机。

A processor should be though of as a switch. On or off. 1 or 0 (binary) or true or false (Boolean). These values are in turn converted to machine code that is the building blocks of what we do digitally.
处理器应该作为一个开关。开或关 1 或 0（二进制）或 true 或 false（布尔）。这些值反过来又被转换为机器代码，这是我们数字化工作的基石。

That being said- it is very complex to come up with a program (that’s all you can really do with a computer) that can intelligently learn in the same way humans or animals can. Lets go back to the definition of intelligence-
话虽如此 - 想出一个可以像人类或动物一样智能学习的程序（这就是你真正能用计算机做的全部事情）是非常复杂的。让我们回到智能的定义——

The ability to acquire and apply knowledge and skills.
获取和应用知识和技能的能力。

So how generally (don’t shoot me programmers!) this applied in computer science in two ways.
那么，这在计算机科学中以两种方式普遍应用（不要向我开枪！）。

Machine Learning 机器学习

Machine learning is an application that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of computer programs that can access data and use it learn for themselves.
机器学习是一种应用程序，它使系统能够自动从经验中学习和改进，而无需明确编程。机器学习侧重于开发可以访问数据并使用数据自行学习的计算机程序。

So basically in a nutshell that means that decision making is left to retrospective decisions and applied generally in a probability model. So, if the machine has come across a decision before, and taken the perceived correct route- it calculates a probability based on previous decisions it has made before and off it goes. I suppose this can be perceived as 'learning'.
所以基本上，简而言之，这意味着决策留给回顾性决策，并普遍应用于概率模型。因此，如果机器之前遇到过决策，并采取了感知正确的路线，那么它就会根据之前做出的决策来计算概率，然后就可以开始了。我想这可以被视为“学习”。

I have applied this in our flagship Smart Reader software here in Sonix, and struggled a lot with the younger programmers who want to call that A.I. Its not.
我已经将其应用于我们在 Sonix 的旗舰 Smart Reader 软件中，并与想要称其为 AI 的年轻程序员进行了很多斗争。

A.I. should mimic human decisions. When was the last time you calculated a probability when you were deciding on food in a restaurant? As much as a nerd as I am, I have never done that.
人工智能应该模仿人类的决定。您上一次在决定餐厅的食物时计算概率是什么时候？尽管我是个书，但我从来没有这样做过。

However, this model lends itself very well to a lot of software, but is called A.I. This makes me cringe a little. Sorry guys!
但是，这种模型非常适合许多软件，但称为 AI。这让我有点畏缩。对不起，伙计们！

Self Modifying Code 自修改代码

This is faily self explanatory. Code, that writes code. We see this a lot in viruses. A virus that probes for weakness and then mutates its code to do something bad (or good).
这是不言自明的。Code，编写代码。我们在病毒中经常看到这种情况。一种探测弱点，然后改变其代码以做坏事（或好事）的病毒。

This type of model is used extensively in the scientific community. I have also used this in software in Sonix, alongside machine learning. This to me is pretty close in our industry to A.I. But not quite.
这种类型的模型在科学界中被广泛使用。我还在 Sonix 的软件中将其与机器学习一起使用。对我来说，这在我们的行业中与 AI 非常接近，但并不完全是。

Again when was the last time you decided to change your thought pattern to decide to like food you never liked before and then ordered it in a restaurant?
再说一次，你上一次决定改变你的思维模式，决定喜欢你以前不喜欢的食物，然后在餐厅点餐是什么时候？

Turing Test 图灵测试

So as previously discussed, Alan Turning came up with a test for A.I., as he believed it would not happen in his lifetime (I believe he was right).
所以如前所述，Alan Turning 想出了一个人工智能测试，因为他认为这在他的有生之年不会发生（我相信他是对的）。

Rather than trying to determine if a machine is thinking, Turing suggests we should ask if the machine can win a game, called the "Imitation Game"( Not to be confused with the film about his code breaking work in Bletchley Park).. The original Imitation game that Turing described is a simple party game involving three players. Player A is a man, player B is a woman and player C (who plays the role of the interrogator) can be of either sex. In the Imitation Game, player C is unable to see either player A or player B (and knows them only as X and Y), and can communicate with them only through written notes or any other form that does not give away any details about their gender. By asking questions of player A and player B, player C tries to determine which of the two is the man and which is the woman. Player A's role is to trick the interrogator into making the wrong decision, while player B attempts to assist the interrogator in making the right one.
与其试图确定机器是否在思考，不如问问机器是否能赢得一场叫做“模仿游戏”的游戏（不要与他在布莱切利公园的密码破译工作的电影相混淆）..图灵描述的原始 Imitation 游戏是一个涉及三个玩家的简单派对游戏。玩家 A 是男性，玩家 B 是女性，玩家 C（扮演审讯者的角色）可以是任何性别。在模仿游戏中，玩家 C 无法看到玩家 A 或玩家 B（并且只知道他们是 X 和 Y），只能通过书面笔记或任何其他不会泄露他们性别详细信息的形式与他们交流。通过向玩家 A 和玩家 B 提问，玩家 C 试图确定两者中哪一个是男性，哪个是女性。玩家 A 的角色是欺骗审讯者做出错误的决定，而玩家 B 则试图帮助审讯者做出正确的决定。

Turing proposes a variation of this game that involves the computer: '"What will happen when a machine takes the part of A in this game?" Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman? These questions replace our original, 'Can machines think?"' So the modified game becomes one that involves three participants in isolated rooms: a computer (which is being tested), a human, and a (human) judge. The human judge can converse with both the human and the computer by typing into a terminal. Both the computer and human try to convince the judge that they are the human. If the judge cannot consistently tell which is which, then the computer wins the game.

Essentially, can you be fooled by a machine into believing it is a person?

Alexa frightened me once, by saying something in the middle of the night but I think that may have been my T.V. setting her off.

An amusing test here at Sonix by Ryan Jones (Developer, Sonix Software)

One of my devs here in Sonix accidentally did a test once that I think was pretty important, at least to me. Ryan Jones came to me with no experience, and wanted to be a programmer. I set him up with a raspberry pi, and a google assistant hat. I asked him to build me some software that could use both Google assistant and the software version of Alexa to control a robot. The result was pretty funny.

When both were turned on, and call names were the same, they have a conversation.

The conversation begins fairly normally, but pretty quickly goes into the absurd, and at the end is in Spanish. Interestingly enough when I heard that I asked him to run this 3 times over. Each time the conversation is exactly the same. So, with readily available kit, you can’t really mimic a normal human conversation. Not without a human to ask the questions. Don’t get me wrong I certainly am not comparing myself to the great Turing’s research!

Maybe this article is now delving into the absurd itself, so let me finish it up.

When companies say they use A.I. in their applications, unless they are keeping a huge secret – I don’t believe it. I believe the terminology and Turing’s definition should be applied or maybe call it something else. I know it sounds so sexy to say that and application is powered by A.I., but lets dial it back a bit. We are not in Star Trek and we don’t have that yet.

Lastly, Alan Turing was instrumental in breaking the Enigma code, saving countless lives in World War 2 and was one of the finest minds in the field. He was convicted of homosexuality and given chemical treatment. He had his passport taken away and committed suicide. Through own stupidity and misunderstanding we lost out on research that may well have given us A.I. Lets never do that again. 

Click on the robot for more information about Sonix Software.